Bulimia nervosa (BN) is a disorder characterized by uncontrolled overeating, self-induced vomiting or other extreme forms of weight control, and dysfunctional attitudes to body shape and weight, affecting some 2.0% of young women. Associated psychopathology includes depression, anxiety, and personality disorders. Although its causes have not been identified, BN can be treated reasonably effectively with psychological therapy. Several studies have established that cognitive-- behavior therapy (CBT) is an effective treatment with between 50% and 60% of patients ceasing to binge eat and purge as a result of treatment. A recent finding suggests that interpersonal psychotherapy (IPT) may produce equally favorable effects. The latter finding has significant therapeutic implications and is also of major theoretical importance because IPT, which is conceptually and procedurally distinct from CBT, does not focus on eating behavior or attitudes. The overall aim of the proposed study is to compare the efficacy of CBT and IPT, and so probe the nature of BN. More specifically, the study aims to determine how much change occurs, on what features of BN, in what sequence, in whom, and via what mechanisms. Two hundred women with BN will be randomized to receive individual treatment with either CBT or IPT at two treatment sites. A dual-site study is required to recruit the large sample needed to provide sufficient statistical power to accomplish the study aims, and to establish generalizability of findings. Well-trained therapists will administer both treatments using detailed manuals under continual monitoring by a third quality control center. Binge eating, purging, and other features of BN will be comprehensively assessed pretreatment, at the midpoint of treatment, posttreatment (after five months of therapy), and at 4, 8, and 12 month follow-up. The study will provide valuable data about BN and its treatment, and have broader significance for the investigation of psychosocial therapies.